Quantum key distribution (QKD) methods and systems have been developed which enable two parties to share random data in a way that has a very high probability of detecting any eavesdroppers. This means that if no eavesdroppers are detected, the parties can have a high degree of confidence that the shared random data is secret. QKD methods and systems are described, for example, in U.S. Pat. No. 5,515,438, U.S. Pat. No. 5,999,285 and GB 2427317 A.
Whatever particular QKD system is used. QKD methods typically involve QKD transmitting apparatus 10 (see FIG. 1 of the accompanying drawings) using a QKD transmitter 12 to send a random data set provided by random source 11, over a quantum signal channel 5 to a QKD receiver 22 of QKD receiving apparatus 20. The QKD transmitting and receiving apparatus 10, 20 then respectively process the data transmitted and received via the quantum signal channel in respective processing sub-systems 13, 23 thereby to derive a common subset m of the random data set. This processing is done with the aid of messages exchanged between the processing sub-systems 13, 23 via an insecure classical communication channel 6 established between classical channel transceivers 14 and 24 of the transmitting and receiving apparatus 10, 20 respectively. As the quantum signal channel 5 is a noisy channel, the processing of the data received over that channel includes an error correction phase that relies on messages exchanged over the classical channel 6.
In some QKD systems, the quantum signal is embodied as a stream of randomly polarized photons sent from the transmitting apparatus to the receiving apparatus either through a fiber-optic cable or free space; such systems typically operate according to the BB84 quantum coding scheme (see for example C. H. Bennett and G. Brassard “Quantum Cryptography: Public Key Distribution and Coin Tossing”, Proceedings of IEEE International Conference on Computers Systems and Signal Processing, Bangalore, India, December 1984, pp 175-179).
In such systems, the QKD transmitter 12 provides the optical components for selectively polarizing photons, and the QKD receiver 22 provides the optical components for receiving photons and detecting their polarization. Typically, these optical components establish two pairs of orthogonal polarization axes, the two pairs of polarization axes being offset by 45° relative to each other. Conventionally, these two pairs of polarization axes are referred to as vertical/horizontal and diagonal/anti-diagonal respectively. An example QKD transmitter 12 and QKD receiver 22 will now be described with reference to FIGS. 2 and 3 respectively of the accompanying drawings.
The QKD transmitter 12 of FIG. 2 comprises an array of light emitting diodes (LEDs) 15A-D in front of each of which is a respective polarizing filter 16A-16D. Filter 16A polarizes photons emitted from LED 15A vertically, filter 16B polarizes photons emitted from LED 15B horizontally, filter 16C polarizes photons emitted from LED 16C diagonally and filter 16D polarizes photons emitted from LED 16D anti-diagonally. Thus, each photon in the stream of photons coming away from the filters 16A-D, is polarized in one of four directions, these directions corresponding to two pairs of orthogonal polarization axes at 45° to each other. A fibre optic light guide 17 conveys the polarized photons out through a lens via a narrow band pass frequency filter 18 (for restricting the emitted photons to a narrow frequency range, typically plus or minus 1 nm), and a spatial filter 19 (for restricting the emitted photons to a narrow frequency range and for obfuscating the characteristics of the particular LED that fired). An attenuation arrangement, not specifically illustrated, is also provided is to reduce the number of photons emitted; the attenuation arrangement may simply be an attenuating filter placed near the other filters or may take the form of individual power control circuits for regulating the power fed to each LED 15A to 15D when pulsed. Without the attenuation arrangement the number of photons emitted each time a LED is pulsed at normal levels would, for example, be of the order of one million; with the attenuation arrangement in place, the average emission rate is 1 photon per 10 pulses. Importantly this means that more than one photon is rarely emitted per pulse.
The FIG. 3 QKD receiver 22 comprises a lens 25, a quad-detector arrangement 30, and a fibre optic light guide 26 for conveying photons received through the lens 25 to the quad-detector arrangement 30. The quad-detector arrangement 30 comprises a beam splitter 31, a half-wave plate 36 for rotating the polarization of photons by 45°, a first paired-detector unit 32, and a second paired-detector unit 33. The first paired-detector unit 32 comprises a polarization-dependent beam splitter 34 and detectors 37A, 37B; the presence of the beam splitter 34 causes the polarizations detected by the detectors 37A and 37B to be mutually orthogonal. The second paired-detector unit 33 comprises a polarization-dependent beam splitter 35 and detectors 37C, 37D; the presence of the beam splitter 35 causes the polarizations detected by the detectors 37C and 37D to be mutually orthogonal. The polarization rotation effected by half-wave plate 36 causes the polarizations detected by the detectors 37A, 37B to be at 45° to those detected by the detectors 37C, 37D; more specifically, the paired detector unit 33 is arranged to detect horizontal/vertical polarizations whereas the paired detector unit 33 is arranged to detect diagonal/anti-diagonal polarizations.
The beam splitter 31 is depicted in FIG. 3 as half-silvered mirror but can be of other forms such as diffraction gratings. The polarization-dependent beam splitters 34, 35 are, for example, birefringent beam splitters.
Operation of the apparatus of FIGS. 1 to 3 in accordance with the BB84 protocol is generally as follows with the QKD transmitting apparatus 10 and QKD receiving apparatus 20 being conventionally referred to as ‘Alice’ and ‘Bob’ respectively.
Alice randomly generates (using source 11) a multiplicity of pairs of bits, typically of the order of 108 pairs. Each pair of bits consists of a data bit and a basis bit, the latter indicating the pair of polarization axes to be used for sending the data bit, be it vertical/horizontal or diagonal/anti-diagonal. A horizontally or diagonally polarized photon indicates a binary 1, while a vertically or anti-diagonally polarized photon indicates a binary 0. The data bit of each pair is thus sent by Alice over the quantum signal channel 5 encoded according to the pair of polarization directions indicated by the basis bit of the same pair. When receiving the quantum signal from Alice. Bob randomly chooses, by virtue of the action of the half-silvered mirror 31, which paired-detector unit 32, 33 and thus which basis (pair of polarization directions) it will use to detect the quantum signal during each time slot and records the results. The sending of the data bits of the randomly-generated pairs of bits is the only communication that need occur using the quantum channel 5.
Next, Bob sends Alice, via the classical channel 6, partial reception data comprising the time slots in which a signal was received, and the basis (i.e. pair of polarization directions) thereof, but not the data bit values determined as received.
Alice then determines a subset m of its transmitted data as the data bit values transmitted for the time slots for which Bob received the quantum signal and used the correct basis for determining the received bit value. Alice also sends Bob, via the classical channel 6, information identifying the time slots holding the data bit values of m. Bob then determines the data bit values making up the received data. The next phase of operation is error correction of the received data by an error correction process involving messages passed over the classical channel 6; after error correction, Bob's received data should match the data m held by Alice and this can be confirmed by exchanging checksums over the classical channel 6.
Accordingly, the QKD system transmits information by the use of polarized photons or by phase encoding photons. Alice creates a stream of encoded photons and Bob has to detect them. It is usual for these photons to be emitted at a fixed time interval. For example every 100 ns, a photon is emitted by Alice. Bob knows this fact and can therefore make measurements at the appropriate time. A synchronised clock between Alice and Bob can be provided by way of a clock signal sent from Alice to Bob, which can be in the form of a regular light pulse (carrying no information) for example. Such a pulse can be sent at regular intervals, thus enabling Bob to maintain synchronisation with Alice, and determine a rate at which bits are transmitted.